Vol. 1, 447-457, April 1995 Clinical Cancer Research 447

Coexpression of a Multidrug-resistance Gene (MDR1) and Herpes

Simplex Virus Thymidine Kinase Gene as Part of a Bicistronic

Messenger RNA in a Retrovirus Vector Allows Selective

Killing of MDR1-transduced Cells1

Yoshikazu Sugimoto,2 Christine A. Hrycyna,3

Ivan Aksentijevich, Ira Pastan,

and Michael M. Gottesman4Laboratory of Cell Biology [Y. S., C. A. H., I. A., M. M. G.] and

Laboratory of Molecular Biology [I. P.], National Cancer Institute,

National Institutes of Health, Bethesda, Maryland 20892

ABSTRACT

A new retroviral vector, pSXLC/pHa, was constructedto coexpress drug-selectable markers with a second gene of

interest as a part of a bicistronic mRNA in a retroviral

vector using an internal ribosome entry site (IRES) from

encephalomyocarditis virus. This system was used to de-

velop a new retroviral vector pHa-MDR-IRES-TK which

expresses a single mRNA from which translation of the

MDR1 gene is cap dependent and translation of the herpes

simplex virus thymidine kinase gene is IRES dependent. The

pHa-MDR-IRES-TK transfectants showed high levels of

P-glycoprotein expression and multidrug resistance. More

than 95% of the vincristine-resistant cells transfected ortransduced with pHa-MDR-IRES-TK showed hypersensi-tivity to ganciclovir, which selects against cells expressing

herpes simplex virus thymidine kinase. An amphotropic

retrovirus titer of 7.8 x i0�/ml was obtained with this

vector. This safety-modified vector should be useful for

introducing the MDR1 gene into bone marrow cells to pro-

tect normal cells from the toxic effects of cancer chemother-

apy because this vector allows the elimination of cancer cells

that have been unintentionally transduced with the MDR1

vector.

INTRODUCTION

In cancer chemotherapy, there are two major problems tobe overcome. One is the innate or acquired resistance of cancer

Received 9/19/94; accepted 12/7/94.I This investigation has been aided by a grant from The Jane Coffin

Childs Memorial Fund for Medical Research.

2 Recipient of a fellowship from the National Cancer Institute-Japanese

Foundation for Cancer Research Research Training Program. Present

address: Cancer Chemotherapy Center, Japanese Foundation for Cancer

Research, 1-37-1 Kami-Ikebukuro, Toshima-ku, Tokyo 170, Japan.

3 Fellow of The Jane Coffin Childs Memorial Fund for Medical Re-search.

4 To whom requests for reprints should be addressed, at Laboratory ofCell Biology, Building 37, Room 1B22, National Cancer Institute,National Institutes of Health, 37 Convent Drive MSC 4255, Bethesda,

MD 20892-4255.

cells to anticancer drugs, and the other is the toxicity of the

chemotherapeutic drugs to certain normal tissues, such as bone

marrow. The study of the mechanisms of drug resistance in

cancer cells has led to the identification of some of the genes

and gene products that confer drug resistance. In particular, cell

lines showing resistance to multiple drugs such as Vinca alka-

bids, anthracyclines, epipodophyllotoxins, and actinomycin D

have been studied intensively, and one gene responsible for this

form of multidrug resistance, termed MDR1,5 has been identi-

fied and a full-length eDNA of the gene has been cloned and

sequenced (1, 2). The MDRI gene encodes the plasma mem-

brane P-glycoprotein with a molecular mass of 170 kDa. P-

glycoprotein acts as an ATP-dependent efflux pump for various

structurally unrelated natural product antitumor agents (re-

viewed in Ref. 3). The MDR1 eDNA has been shown to confer

multidrug resistance when introduced into drug-sensitive cells

(4, 5).

The MDR1 gene is normally expressed on the biliary

surfaces of hepatocytes, the brush border of the proximal tubules

of kidney, the lumen of small and large intestine, and the

capillary endothelial cells of the brain and testes (3). However,

it is not widely expressed in bone marrow cells with the possible

exception of some CD34-positive cells (6). Lack of protection

by an endogenous transporter may be one of the reasons for the

severe suppression of bone marrow cells by many chemother-

apeutic drugs, which is a major dose-limiting factor in cancer

chemotherapy. Retrovirus-mediated expression of the MDR1

cDNA has been shown to confer multidrug resistance in vivo

when the MDR1-carrying vector is introduced into bone marrow

cells of mice (7, 8). Studies using MDR1-transgenic mice sug-

gest that the expression of the human MDR1 cDNA in bone

marrow cells does not affect the normal function of the bone

marrow cells (9, 10). Therefore, in principle, the MDR1 gene

can be used to protect bone marrow cells from intensive che-

motherapy.

There still remain, however, some potential problems as-

sociated with MDR1 gene transfer into normal bone marrow

cells. One problem is to optimize the efficiency of MDR1 gene

transfer and expression. Using an MDR1 retrovirus, relatively

efficient MDR1 gene transfer is possible, but development of

higher titer retroviruses and an optimized protocol for selection

of the transduced cells in vivo are still required. Another prob-

5 The abbreviations used are: MDR, multidrug-resistance gene; HSV-TK, herpes simplex virus thymidine kinase; IRES, internal ribosomeentry site; TK, thymidine kinase; LTR, long terminal repeat; IC5�,

concentration of drug that inhibits cell growth by 50%; FACS, fluores-

cence-activated cell sorting; HAT, hypoxanthine-aminopterin-thymi-

dine; MCS, multicloning site.

Research. on March 23, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

448 Retroviral Coexpression of MDR1 and HSV-TK

lem is that it is possible to transduce other cells contained within

bone marrow preparations in which the expression of the trans-

duced gene might cause undesirable side effects. In transducing

bone marrow cells of cancer patients with the MDR1 retrovirus,

if the bone marrow contains contaminating cancer cells and the

cancer cells receive the MDR1 retrovirus, then multidrug-resis-

tant cancer cells will result. In such cases, treatment protocols

involving drugs related to multidrug resistance will be ineffec-

tive. To eliminate such unintentionally transduced cells, intro-

duction of a negative drug-selectable marker (a ‘ ‘suicide’ ‘ gene)

that confers hypersensitivity to a certain drug would be valuable.

The HSV-TK gene acts as a suicide gene both in vitro and

in vivo. Cells that express HSV-TK are hypersensitive to the

nucleoside analogue ganciclovir because HSV-TK can phos-

phorylate ganciclovir more efficiently than the endogenous TK

of mammalian cells. Ganciclovir treatment was reported to be

effective in vivo against rat cerebral gliomas transduced with the

retrovirus carrying HSV-TK (ii, 12). Multidrug-resistant cells

do not show cross-resistance to ganciclovir. Therefore, the

HSV-TK gene would appear to be a suitable gene to coexpress

with the MDR1 gene to eliminate cancer cells transduced with

the MDR1 vector. HSV-TK can also confer HAT resistance in

vitro when it is introduced into TK-deficient cells such as Ltk

There are three possible strategies for coexpression of the

MDR1 gene with the HSV-TK gene. The first strategy is to

utilize two independent promoters (e.g., retroviral LTR and

another promoter). However, this does not guarantee the expres-

sion of both genes. Most transduced cells express only one gene

because expression of one gene may suppress the activity of the

second promoter (13). The second strategy is to engineer a

chimeric bifunctional protein between P-glycoprotein and HSV-

TK. This approach has been shown to be successful when a

second gene product, such as adenosine deaminase, is connected

to the carboxyl terminus of P-glycoprotein (14, 15). This strat-

egy guarantees the coexpression of the two gene products;

however, the activity of chimeric P-glycoprotein as a drug

transporter is less than the intact P-glycoprotein and some gene

products may not function when tethered to the inside of the

plasma membrane. The third strategy is to use an IRES isolated

from a picornavirus such as encephalomyocarditis virus (16-

18). In this construct, a single mRNA is transcribed under the

control of an upstream promoter, and two gene products are

translated independently from the mRNA. The first open read-

ing frame is translated in a cap-dependent fashion, and the

second is translated under control of the IRES.

We have developed a new retroviral vector system,

pSXLC/pHa, in which the drug-selectable genes such as the

MDR1 gene or the HSV-TK gene is translated under control of

the IRES (19). To express the MDR1 gene and the HSV-TK

gene together, we constructed a new MDR1-retrovirus vector

pHa-MDR-IRES-TK that allows simultaneous translation of the

cap-dependent MDRJ gene and the IRES-dependent HSV-TK

gene.

MATERIALS AND METHODS

Cell Culture and Drug Sensitivity Assay. The edo-

tropic retrovirus packaging cell line �I’-cre, the amphotropic

retrovirus packaging cell line �P-crip (20), and the mouse fibro-

I LTRF_1MDRJ_H LTRIpHaMDR

I LTR � IRES � TK xi�:�_�1LTR

pHa-MCS-IRES-TK

I LTRHIMD4IRESI TK1__-1L1�RIpHa-MDR-IRES-TK

Fig. I Structure of pHaMDR, pHa-MCS-IRES-TK, and pHa-MDR-IRES-TK retrovirus. Drawing is not to scale. LTR, LTR of Harvey

murine sarcoma virus; MDR, human multidrug resistance gene MDR1;

TK, herpes simplex virus thymidine kinase gene; S, Sacli; Xh, XhoI;

MCS, SacII-BamHI-BglII-SacI-XbaI-SalI; N, NcoI; Xb, XbaI.

blast cell line NIH3T3 were cultured in DMEM supplemented

with 10% calf serum. The murine TK-deficient cell line Ltk

and the amphotropic retrovirus packaging cell line PA317 (21)

were grown in DMEM supplemented with 10% fetal bovine

serum. The sensitivities of the cultured cell lines to drugs were

evaluated by the inhibition of cell growth after incubation at

37#{176}Cfor 6 days in the presence of various concentrations of

drugs as described previously (22). Cell numbers were deter-

mined in a Coulter counter and the IC50 was calculated.

Construction of Vectors. The construction of the

pSXLC/pHa retrovirus system was described previously (19).

The plasmid pSXLC-TK has the entire open reading frame of

HSV-TK DNA downstream from the IRES sequence and 6

unique sites (SaclI, BamHI, BglII, Sad, XbaI, and SalI) for the

cloning of another gene upstream from the IRES. To insert the

MDR1 eDNA into pSXLC-TK, we made pSXbaMDR, which

contains MDR1 eDNA with 5’-SacII and 3’-XbaI sites in a

pGEM2-derived vector, then subcloned the SacII-XbaI-digested

MDR1 eDNA between the SaclI and XbaI sites of pSXLC-TK

(pSXLC-MDR-TK). The pSXLC-MDR-TK insert was isolated

after SacII-XhoI digestion and transferred into the pHa retroviral

vector (pHa-MDR-IRES-TK). pHaMDR, which carries the

wild-type MDR1 eDNA in the pHa retrovirus vector (5, 23), was

used as a control MDR1 vector. pHa-MCS-IRES-TK, which

carries the insert of pSXLC-TK, was also used as a control (19).

Structures of retrovirus constructs used in this study are sum-

marized in Fig. 1.

DNA Transfection. Transfection was carried out using

the calcium phosphate coprecipitation method (24). Recipient

cells were plated at 5 x i0� cells/100-mm dish on day 1 and

transfected with 20 jig of the expression plasmid DNA on day

2. Cells were exposed to the DNA precipitate until day 3 when

the medium was aspirated and fresh medium was added. On day

4, the cells were split at 1:10 or 1:100. The cells were selected

either in vincristine (25 ng/ml for ‘I’-cre, ‘P-drip, and NIH3T3;

30 ng/ml for PA317 and 35 ng/ml for Ltk) or HAT medium

(100 p.M hypoxanthine, 0.4 p.M aminopterin, and 16 p.M thymi-

dine) on day 5, if necessary.

Retrovirus Transduction. To examine the production ofretrovirus, retrovirus-producing cells were plated on day 1 at

Research. on March 23, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

Clinical Cancer Research 449

2 x 106 cells/100-mm dish. On day 2, the medium of the

packaging cell culture was changed, and the recipient cells were

plated at 3 x iO� cells/100-mm dish in medium containing

2 p.g/ml polybrene (Aldrich, Milwaukee, WI). On day 3, the

retrovirus-containing supernatant was collected, passed through

a 0.45-jim pore filter to remove cells and debris, and added to

each dish of recipient cells. On day 5, the medium was removed,

and fresh medium containing drugs was added at appropriate

concentrations as needed. To estimate the retrovirus titer, the

medium was removed and the colonies were stained with 0.5%

methylene blue dissolved in 50% methanol on days 11-13.

Protein Expression Analysis. To examine the expres-

sion of human P-glycoprotein on the cell surface of mouse

transfectants, FACS analysis was used in which cells were

reacted with a human P-glycoprotein-specific mAb MRK16 (25,

26). Cells (106) harvested after trypsinization were washed and

incubated with MRK16 (5 p.g/106 cells), washed twice, and

incubated with fluorescein-conjugated goat anti-mouse IgG

(1:10 diluted; Jaxon Immunoresearch Lab., West Grove, PA).

The cells were washed twice and the fluorescence staining level

was analyzed using a FACSort (Becton Dickinson FACS Sys-

tem, San Jose, CA).

Nucleic Acid Isolation and Analysis. Total RNA was

isolated from cultured cells by acid-guanidinium-thiocyanate-

phenol-chloroform extraction (Stratagene, Inc., La Jolla, CA;

Ref. 27). Cells were first homogenized by passage through a

20-gauge needle after removal from the dish by direct addition

ofthe guanidinium thiocyanate. For Northern blot hybridization,

RNA was resolved on a 1% formaldehyde-agarose gel and

subsequently transferred to nitrocellulose (BA83; Schleicher &

Schuell) (28). The blot was probed with a random-primed 0.8-

kilobase MDR1 probe obtained from an EcoRI-HindIII diges-

tion of MDR1 cDNA (29). The blot was subsequently stripped

and reprobed with a random-primed 1.2-kilobase NcoI-XhoI

fragment from pSXLC-TK containing the entire open reading

frame for HSV-TK. Hybridization and washing conditions were

as described previously (30).

RESULTS

We have been using the pHaMDR retroviral vector which

uses the promoter of the Harvey murine sarcoma virus LTR to

drive expression of the MDR1 gene in mammalian cells (5, 31).

In a previous study we reported the generation of a retroviralvector system pSXLC/pHa that uses the IRES sequence to

coexpress drug-selectable genes with the gene of interest (19).

To coexpress the HSV-TK gene as a suicide gene in the pHa

vector, we made pSXLC-TK with the entire open reading frame

of the HSV-TK gene downstream from the IRES and six unique

sites (Sad!, BamHI, BglII, Sad, XbaI, and Sail) for the cloning

of another gene (MDR1) upstream from the IRES. This

IRES-TK construct was shown to be effective when the insert of

pSXLC was transferred into the pHa vector (pHa-MCS-IRES-

MDR) and introduced to Ltk cells (19). In the current study,

we inserted the MDR1 cDNA into pSXLC-TK upstream from

the IRES (pSXLC-MDR-TK) and transferred the whole insert

into the pHa vector. The resulting construct was termed pHa-

MDR-IRES-TK (Fig. 1).

Table I Transfection efficiency of pHa-MDR-IRES-TK and

pHaMDR

Cell Vector

Transfection efficiency”

Vincristine” HAT

‘I’-cre

Ltk

pHa-MDR-IRES-TKpHaMDRpMClneopolyApHa-MDR-IRES-TKpHaMDRpMCineopolyA

3.2 X i0�8.4 x i0�

02.0 X i0�3.2 x i0�

0

nd”ndnd

7.6 X 10�00

nies.

a Calculated probability of the emergence of drug-resistant cob-

/) The transfected cells were selected with vincristine at 25 ng/ml

for �1’-cre cells and 35 ng/ml for Ltk cells.C The transfected cells were selected in HAT medium.

d nd, not determined.

Both MDR1 and HSV-TK Genes Are Expressed in CellsTransfected with pHa-MDR-IRES-TK. The ecotropic retro-

virus-packaging cells ‘P-cre and the murine TK-deficient cells

Ltk were transfected with the retroviral expression constructs,

pHa-MDR-IRES-TK and pHaMDR, and selected with vincris-

tine as described in ‘ ‘ Materials and Methods. ‘ ‘ Ten days after

transfection, approximately 1 00 vincristine-resistant colonies

per dish were observed in pHa-MDR-IRES-TK-transfected

‘I’-cre cells when transfected cells were split 1:100 on day 4,

giving a transfection efficiency of 3.2 X i0�. Transfection

efficiencies of pHa-MDR-IRES-TK and pHaMDR are summa-

rized in Table 1. As shown in Table 1, both the pHa-MDR-

IRES-TK and pHaMDR vectors efficiently transformed ‘I’-cre

and Ltk cells to vincristine resistance. No vincristine-resistant

colonies were found in pMC1 neopoly(A) (a G418-resistant

plasmid obtained from Stratagene)-transfected cells. This result

clearly shows that the MDR1 gene is expressed in pHa-MDR-

IRES-TK-transfected cells. Transfection efficiencies of pHa-

MDR-IRES-TK were 2-3-fold less than those of pHaMDR(Table 1). Higher numbers of vincristine-resistant colonies were

obtained in ‘F-cre cells than in Ltk cells. The Ltk cells

transfected with the MDR1 vectors were also selected with HAT

medium (Table 1). When the pHa-MDR-IRES-TK-transfected

Ltk cells were selected with HAT medium, approximately 190

HAT-resistant colonies/dish were observed (transfection effi-

ciency, 7.6 x 10�). No HAT-resistant colonies were found

among pHaMDR-transfected cells or pMC1 neopoly(A)-trans-

fected cells. These results indicate that pHa-MDR-IRES-TK

confers both vincristine resistance and HAT resistance in Ltk

cells. Three to four times as many colonies were obtained by

HAT selection than by vincristine selection.

Expression of a single bicistronic mRNA species encod-

ing both P-glycoprotein and HSV-TK was confirmed by North-

em blot analysis performed on RNA isolated from 1-LAT-resis-

tant transfectants using MDR1-and HSV-TK-specific cDNA

fragments (Fig. 2). In the RNA from Ltk cells transfected with

the pHa-MDR-IRES-TK construct, both probes revealed a sin-

gle transcript of approximately 10-10.5 kilobases correspond-

ing to the predicted size (Fig. 2, Lanes 4). Using the MDR1

probe, no signal was detected in the RNA from Ltk cells

transfected with pHa-MCS-IRES-TK (Fig. 14, Lane 2), but the

Research. on March 23, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

A B12 34 12 34

- on

�1V -28S� 0 #{149} -18S

450 Retroviral Coexpression of MDRI and HSV-TK

the presence of vincristine (15 ng/ml, ViS or 30 ng/ml, V30) or

HAT medium. The results shown in Table 2 clearly demonstrate

that both the MDR1 and HSV-TK gene are expressed together in

most of the transfectants. All of the HAT-resistant clones (20/

20) showed more resistance to vincristine than the parental

Ltk , although some clones showed only marginal levels of

resistance. On the other hand, 19 of 20 vincristine-resistant

clones showed HAT resistance. Only one clone (termed L39)

showed HAT sensitivity. The pHa-MDR-IRES-TK clones se-

lected with vincristine appeared to have higher levels of vine-

ristine resistance than the transfectants selected with HAT me-

dium.

Fig. 2 Northern blot analysis of KB-V1 cells and transfectants with

pHa-MCS-IRES-TK, pHaMDR, or pHa-MDR-IRES-TK. In A, total

RNA was isolated from cells, fractionated on a 1% formaldehyde-

agarose gel (20 jig/lane), and transferred to a nitrocellubose membrane

as described in ‘ ‘ Materials and Methods. ‘ ‘ The membrane was then

probed with a 0.8-kibobase MDR1 probe. In B, the membrane was

subsequently stripped and reprobed with a 1.2-kilobase probe for HSV-

TK. A and B, Lane 1, NIH3T3 cells transfected with pHaMDR and

selected in 60 ng/ml colchicine; Lane 2, Ltk cells transfected with

pHa-MCS-IRES-TK and selected in HAT medium; Lane 3, multidrug-

resistant KB-Vt cells grown in I pg/ml vinhlastine; Lane 4, Ltk cellstransfected with pHa-MDR-IRES-TK and selected in HAT medium.

Rig/it, positions of the origin (on) and 185 and 285 species. Arrow,

expected position of the pHa-MDR-IRES-TK transcript.

correct size species of approximately 7 kilobases was detected

in these same cells using the HSV-TK probe (Fig. 2B, Lane 2).

As additional controls, the MDR1 probe revealed the expected

size transcripts of approximately 8 kibobases and 4.4 kilobases

in NIH3T3 cells transfected with pHaMDR (5) and in KB-V1

cells (32), respectively (Fig. 2A, Lanes I and 3), but as predicted

no signal was observed using the HSV-TK probe (Fig. 2B, Lanes

I and 3).

Expression of P-glycoprotein on the cell surface was ex-

amined by FACS with a human P-glycoprotein-specific mAb

MRK16 (25). Mixed populations of vincristine-resistant �1’-cre

cells transfected with pHa-MDR-IRES-TK and selected with 25

ng/ml vincristine showed significant expression of the human

P-glycoprotein (Fig. 3A). The complete shift of the fluorescence

peak of the transfectants suggests that all of the transfectants

selected with vincristine express human P-glycoprotein. The

pHa-MDR-IRES-TK transfected ‘I’-cre cells showed slightly

lower expression of P-glycoprotein than the pHaMDR-trans-

fected cells (Fig. 3B). Mixed populations of the pHa-MDR-

IRES-TK-transfected Ltk cells selected with vincristine (Fig.

3C) or HAT medium (Fig. 3E) also showed significant expres-

sion of human P-glycoprotein. The expression of P-glycoprotein

in the vincristine-resistant cells was higher than the expression

in the HAT-resistant cells. The shift in the fluorescence peak of

the HAT-resistant transfectants suggests that all of the transfec-

tants selected with HAT medium express human P-glycoprotein.

Again, the pHa-MDR-IRES-TK-transfected Ltk cells showed

slightly lower expression of P-glycoprotein than the pHaMDR-

transfected cells (Fig. 3, C and D).

To determine whether the two genes are coexpressed in

clonal cell lines, we randomly isolated 20 vincristine-resistant

clones and 20 HAT-resistant clones from pHa-MDR-IRES-TK-

transfected Ltk cells. Growth of each clone was examined in

Sensitivities to vincristine, ganciclovir, and HAT medium

of pHa-MDR-IRES-TK-transfected Ltk cell clones were de-

termined and are shown in Table 3. In HAT medium, aminop-

term acts as a toxic substance and hypoxanthine and thymidine

are needed to protect TK-expressing cells from the toxicity of

aminopterin. Therefore, we determined the ICS() values of am-

inopterin in the presence of hypoxanthine (100 jiM) and thymi-

dine (16 jiM) as a measure of sensitivity to HAT medium.

Growth of most pHa-MDR-IRES-TK transfectant clones was

not inhibited at 400 nM aminopterin, which is the concentration

of this drug in HAT medium. The ICSt) values for ganciclovir in

the pHa-MDR-IRES-TK-transfected clones (except for clone

L39) were 10-40 nM, whereas those of the parental Ltk cells

or control MDR1 transfectants were 16-20 p.M. The ICSt) value

for ganciclovir in clone L39 was 8.6 jiM. This result indicates

that ganciclovir shows very high selective toxicity (approxi-

mately 103-fold) against cells expressing the HSV-TK gene.

Ganciclovir has been shown to be effective in vivo against

HSV-TK-transduced cells (1 1, 12). Therefore this strategy

should be useful to eliminate MDR1-expressing tumor cells that

have been unintentionally transduced with the pHa-MDR-

IRES-TK vector.To examine the probability that cells expressing the MDR1

gene would not express the HSV-TK gene, we determined the

proportion of ganciclovir-resistant cells in the pHa-MDR-IRES-

TK-transfected, vincristine-resistant cell population. ‘I’-cre cells

transfected with pHa-MDR-IRES-TK were treated with vincris-

tine (25 ng/ml) for 7 days and subsequently cultured in drug-free

medium for 2 days. The cells were plated onto 100-mm dishes

at various concentrations of the viable cells and treated with S

jiM ganciclovir. When i04 cells were plated, approximately 160

ganciclovir-resistant colonies were observed. This result shows

that pHa-MDR-IRES-TK can confer hypersensitivity to ganci-

clovir although the recipient cells express endogenous TK.

Since the plating efficiency was 65% (130 colonies from 200

cells in drug-free medium), the frequency of ganciclovir-resis-

tant cells in the pHa-MDR-IRES-TK-transfected ‘I’-cre cells

was 160/0.65 x i0’� 0.025 (Table 4). These ganciclovir-

resistant clones were pooled after trypsinization and tested for

MDR1 expression.

As shown in Fig. 4, A and B, almost 100% of the ganci-

clovir-resistant cells were shown to express human P-glycopro-

tein. We also did a similar experiment using Ltk cells. The

pHa-MDR-IRES-TK-transfected Ltk cells were selected with

vincristine (35 ng/ml), plated, and treated with S p.M ganciclovir

as described above. Approximately 120 ganciclovir-resistant

colonies were found from i0� cells (Table 4). Since the plating

Research. on March 23, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

A B

C D

E

Fluorescence Intensity

ii

Fluorescence Intensity

Clinical Cancer Research 451

-�

E

za)

0

a)

Ezci)

0

a)

E

za)

0

Fig. 3 FACS analysis of the transfectants with pHa-MDR-IRES-TK or pHaMDR. Cells (106) were harvested after trypsinization, stained with theanti-P-glycoprotein mAb MRK16, washed, and stained with fluorescein-conjugated antimouse IgG. The fluorescence was analyzed by FACS. A,

‘I’-cre cells (left) and the pHa-MDR-IRES-TK-transfected ‘I’-cre population selected with vincristine (right); B, ‘11-crc cells (left) and the

pHaMDR-transfected ‘I’-cre population selected with vincristine (right); C, Ltk cells (left) and the pHa-MDR-IRES-TK-transfected Ltk population

selected with vincristine (right); D, Ltk cells (left) and the pHaMDR-transfected Ltk population selected with vincristine (rig/li). E, Ltk cells (left)

and the pHa-MDR-IRES-TK-transfected Ltk population selected with HAT medium (right).

efficiency was 85%, the frequency of ganciclovir-resistant

cells in the pHa-MDR-IRES-TK-transfected Ltk cells was

120/0.85 x i04 = 0.014 (Table 4). These ganciclovir-resistant

Ltk clones were pooled after trypsinization and tested for HAT

sensitivity and the MDR1 expression. When 5 X i04 cells of the

pHa-MDR-IRES-TK-transfected, ganciclovir-resistant popula-

tion was plated and selected with HAT medium, no HAT-

resistant colonies appeared, suggesting that these ganciclovir-

resistant cells do not express the foreign HSV-TK gene. On the

other hand, almost 100% of the ganciclovir-resistant cells were

Research. on March 23, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

452 Retroviral Coexpression of MDR1 and HSV-TK

Table 2 Growth of pHa-MDR-IR ES-TK-transfected Ltk clones n the prese nce of vincristine or HAT

Cell growth (% of control)

HAT-selected clone Vincristine-selected clone

Clone V15” V30” HATClone V15” V30” HAT

LOlL02

L03

L04LOSLO6

L07

L08

L09LII)

Lll

Ll2

L13

L14

LlS

Ll#{244}

Ll7

L18

Ll9

L20

9490

74

78

8088

77

63

8057

48

71

56

43

38

43

51

30

55

33

85 96

60 95

57 101

54 102

49 109

45 97

40 97

35 98

22 9822 10022 102

19 100

19 105

19 95

17 101

16 103

13 97

12 93

9 88

9 101

L21

L22

L23

L24

L25

L26

L27

L28

L29

L30

L31

L32

L33

L34

L35

L36

L37

L38

L39

L40

93 80

103 78

1(X) 74

98 68

90 62

85 55

70 55

84 49

80 48

72 41

77 40

88 37

76 36

70 35

61 35

67 32

69 24

58 21

52 15

53 10

100101

100

98

103

98

98

101

102

95

101

102

103

97

92

105

100

95

0

85

Parental Ltk�

pHaMDR transfectant

pHa-MCS-IRES-TKtransfectant

10 4

105 90

12 5

1

0

95

a Cell growth (percentage of control) in the presence of 15 ng/ml vincristine.1� Cell growth (percentage of control) in the presence of 30 ng/ml vincristine.

( Cell growth (percentage of control) in HAT medium.

shown to express human P-glycoprotein (Fig. 4, C and D),

suggesting that these cells did not have functional HSV-TK, but

the MDRI gene was expressed. As a control experiment, the

pHa-MDR-IRES-TK-transfected cells were selected with HAT

medium. When the HAT-resistant population was treated with 5

jiM ganciclovir, no colonies were observed as expected. These

results suggest that 1-3% of the pHa-MDR-IRES-TK-trans-

fected cells may express the MDR1 gene alone and show resis-

tance to ganciclovir.

Retroviral Transduction of pHa-MDR-IRES-TK. To

test the ability of pHa-MDR-IRES-TK to be packaged as retro-

virus, culture supernatants of vincristine-resistant, pHa-MDR-

IRES-TK-transfected ‘I’-cre cells (mixed population) were

added to the culture of amphotropic retrovirus-packaging lines

PA317 or ‘P-crip. After retrovirus transduction and subsequent

vincristine selection, many drug-resistant colonies were oh-

tamed, indicating that retrovirus carrying the MDR1 gene ex-

isted in the supernatants. The retrovirus titer produced by mixed

populations of ‘I’-cre cells transfected with pHa-MDR-IRES-TK

was 1.8 x 103/ml when PA317 cells were used as recipient

cells, whereas the retrovirus titer produced by a mixed popula-

lion of ‘P-cre cells transfected with pHaMDR was 7.1 X 103/ml.

The retrovirus titers of mixed populations of PA317 or

‘V-drip cells transduced with pHa-MDR-IRES-TK were exam-

med using Ltk and NIH3T3 cells as recipients (Table 5).

PA317 cells produced higher titers of retrovirus on the average

than ‘V-drip cells. Higher numbers of vincristine-resistant cob-

nies were obtained in NIH3T3 cells than in Ltk cells. Retro-

virus titers of pHa-MDR-IRES-TK were 2-3-fold less than

Table 3 Sensitivity of pHa-MDR-IRES-TK-transfected Ltk clones

to vincristine, ganciclovir, and aminopterin

ICS,) values to

Vincristine Ganciclovir Aminopterin

Cell lines (ng/ml) (nM) (nM)

Ltk 5.8 17,000 6.0

pHaMDRVincristine-selected clones

LM1 71 16,000 7.2

LM2 62 20,000 5.5

LM3 21 16,000 6.3

pHa-MCS-IRES-TKHAT-selected clones

LIT! 5.7 26 >6,400

LIT2 5.3 13 >6,400LIT3 5.1 40 >6,400

pHa-MDR-IRES-TKHAT-selected clones

LOl 45 13 >6,400

L02 39 27 >6,400

L03 35 13 >6,400L07 25 13 >6,400L10 21 36 >6,400L14 16 11 >6,400L17 12 17 >6,400

L19 1! 40 >6,400

L20 1 1 15 >6,400

Vincristine-selected clones

L2l 43 26 >6,400L25 36 9 >6,400L33 23 32 >6,400

L39 17 8,600 5.2

L40 15 31 >6,400

Research. on March 23, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

Clinical Cancer Research 453

Table 4 Ratio of ganciclovir-resistant cells in the populations of

vincristine-selected pHa-MDR-IRES-TK-transfectants

Cell

Ganciclovir-resistant colonies

Total colonies” No. %

Ltk�

‘I’-cre

1.7 x i0� 250 1.5

8.5 X l0� 120 1.4

4.3 X i03 75 1.71.3 x i0� 280 2.2

6.5 X iO� 160 2.5

3.2 X i03 100 3.1

“ C alculated number of colonies obtained in drug-free medium.

those of pHaMDR. To obtain retrovirus-producing cells with

higher titers, we isolated PA317 or ‘11-drip clones transduced

with pHa-MDR-IRES-TK retrovirus. The highest titer we ob-

tamed from 17 clones of PA317 was 7.8 x 104/ml, and that

from 14 clones of ‘I’-crip was 2.9 X 104/ml. In general, PA317

cells produced 2-5-fold higher titers of retrovirus than ‘P-drip.

The retroviral titers of pHa-MDR-IRES-TK were a few-fold less

than those of pHaMDR. According to previous results, however,

retroviral titers close to 105/ml may be enough to transduce

mouse bone marrow cells for animal experiments (7, 8). It

seems likely that further screening could result in clones with

higher supernatant retrovirus titers.

As described above, we observed a few-fold difference in

the number of pHa-MDR-IRES-TK transfectant colonies be-

tween vincristine selection and HAT selection. This result may

suggest that many transduced cells with low expression of

P-glycoprotein were killed by the vincristine selection. To esti-

mate more accurately the transduction efficiency of the MDR1

retroviruses, we determined the percentages of MRK16-positive

cells in the transduced cell population without drug selection.

Ltk cells were plated at 3 X itT’ cells/100-mm plate on day 1.

The culture supernatants (1 ml or 2.5 ml) of the retrovirus-

producing PA317 cells (mixed population) were added to the

Ltk cells on day 2. Two independently transduced populations

of PA317 cells producing pHa-MDR-IRES-TK retrovirus were

used in this experiment. On day 4, the medium was removed,

and drug-free medium or HAT medium was added to the cul-

ture. On day 8, cells in the drug-free medium were analyzed by

FACS (Fig. 5, A and B). On the same day, cells selected in HAT

medium were trypsinized and counted in a Coulter counter. The

percentages of MRK16-positive cells and percentages of HAT-

resistant cells (number of cells grown in HAT medium com-

pared to number of cells in drug-free medium) are summarized

in Table 6. As shown in Table 6, the percentages of MRKI6-

positive cells were similar to the percentages of cells that can

grow in HAT medium. As shown in Table 2, HAT medium does

not affect the growth of the pHa-MDR-IRES-TK-transduced cells.

Therefore, the percentages shown in Table 6 represent the actual

transduction efficiency of pHa-MDR-IRES-TK. Since the number

of transfectant colonies obtamed by HAT selection were 2-3-fold

larger than the number obtained by vincristine selection (Table 1),

the actual titer of pHa-MDR-IRES-TK retrovirus may be 2-3-fold

higher than the titers observed after vincristine selection.

Next we compared the transduction efficiency in Ltk cells

with the efficiency in NIH3T3 cells. As described above, the

transfection efficiency of Ltk cells was somewhat lower than

that of ‘P-cre cells (a subline of NIH3T3), and the transduction

efficiency of Ltk cells was lower than that of NIH3T3 (Table

5). The transduction efficiencies of Ltk and NIH3T3 were

obtained from selections using different concentrations of yin-

cristine. Therefore it is not clear what percentage of transduced

cells actually survived during each selection. To compare the

transduction efficiency without drug selection, we examined the

percentages of MRK16-positive cells in the nonselected popu-

lations of transduced cells. This experiment was done at the

same time as the experiment shown in Table 6 as described

above. Ltk and NIH3T3 cells were plated at 3 x i04 cells/

100-mm plate on day 1. The culture supernatants (1 ml or 2.5

ml) of the retrovirus-producing PA3!7 cells (mixed population)

were added to the cells on day 2. On day 4, the medium was

removed, and drug-free medium was added to the culture. On

day 8, cells in drug-free medium were analyzed by FACS (Fig.

5). The percentages of MRK16-positive cells are summarized in

Table 7. As shown in Table 7, the percentages of the cells which

were MRK16 positive were slightly (20-80%) higher in

NIH3T3 cells than in Ltk cells; however, the observed differ-

ence was smaller than the difference in transduction efficiencies

after vincristine selection (2-3-fold, Table 5). These results

suggest that a higher percentage of pHa-MDR-IRES-TK-trans-

duced Ltk cells were killed by the vincristine selection than

that of NIH3T3 cells.

DISCUSSION

In this article we show that it is possible to create a

bicistronic retroviral vector system for coexpression of the hu-

man multidrug resistance gene (MDR1) and a second gene

(HSV-TK). A high percentage of cells express both of these

genes. Since the HSV-TK gene confers sensitivity to ganciclo-

vir, cells transduced with pHa-MDR-IRES-TK can be selec-

tively killed with this drug. In essence, pHa-MDR-IRES-TK is

a safety-modified retroviral vector system in which cells unin-

tentionally selected for multidrug resistance can be killed. Such

a situation could occur with cancer cells during gene therapy

intended to confer drug resistance on normal bone marrow

during chemotherapy. Relatively high titer retroviral superna-

tants (close to 105/ml) can be prepared using the pHa-MDR-

IRES-TK vector.

We have chosen the pHaMDR retrovirus backbone for

these studies since we have had good success in the past using

the promoter of Harvey murine sarcoma virus LTR for the

expression ofthe MDR1 gene in mammalian cells (5, 31). Since

the pHa vector has only two cloning sites, we made a new

subcloning vector pSXLC to introduce various genes into the

retrovirus vector. The pSXLC/pHa system should also prove

useful for cloning other genes upstream from the IRES-TK

sequence because this vector has six unique sites for cloning. In

another study we have demonstrated that this IRES-TK con-

struct was active when the insert of pSXLC-TK was transferred

into the pHa vector (pHa-MCS-IRES-TK) and introduced into

Ltk cells. In this study we introduced the MDR1 gene between

the SacII and XbaI sites of pSXLC-TK (pSXLC-MDR-TK, Fig.

1). The whole insert of pSXLC-MDR-TK was isolated after

Research. on March 23, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

A B

C D

Fluorescence Intensity

454 Retroviral Coexpression of MDR1 and HSV-TK

ci)

Ezci)

0

ci)�0

Ezci)

0

Fig. 4 FACS analysis of the pHa-MDR-IRES-TK transfectants showing resistance to ganciclovir. Cells (1O�’) were harvested after trypsinization,

stained with the anti-P-glycoprotein mAb MRK1o, washed, and stained with fluorescein-eonjugated antimouse IgG. The fluorescence was analyzed

by FACS. A, ‘V-crc cells (left) and the pHa-MDR-IRES-TK-transfeeted ‘l’-ere population selected with vincristine (rig/li); B, ‘V-ere cells (left) and

the pHa-MDR-IRES-TK-transfected ‘V-ere population selected with vineristine and subsequently selected with gancielovir (right); C, Ltk cells (left)

and the pHa-MDR-IRES-TK-transfected Ltk population selected with vincristine (right); D, Ltk cells (left) and the pHa-MDR-IRES-TK-

transfected Ltk population selected with vincristine and subsequently selected with gancielovir (right).

Table 5 Titers of pHa-MDR-IRES-TK and pHaMDR retrovirus

produced by PA317 or ‘V-crip cells determined using Ltk and

NIH3T3 cells as recipient cells

Producer cells/retrovirus

Retrovi rus titer”

Ltk NIH3T3

PA317/pHa-MDR-IRES-TK 1.5 X i0� 3.6 X i0�

PA317/pHaMDR 3.1 X 10� 8.5 X i0�

‘V-erip/pHa-MDR-IRES-TK 0.7 X i0� 2.0 X 10�

---� ‘l’-crip/pHaMDR 1.5 X i0� 2.8 X 10”

“ Calculated number of vincristine-resistant colonies when

NIH3T3 cells transduced with 1-mi supernatants from mixed popula-

tions of retrovirus-producing cells were selected with 25 ng/ml vincris-

tine.

SacII-XhoI digestion and transferred into the pHa retroviral

vector (pHa-MDR-IRES-TK, Fig. 1). Northern blot analysis

demonstrated that this vector allows for expression of a single

bicistronic mRNA species when introduced into cells (Fig. 2).

Additionally, the transfection and transduction experiments us-

ing pHa-MDR-IRES-TK clearly show that the MDR1 gene and

the HSV-TK gene are efficiently expressed in each cell contain-

ing this construct. In a related construction in which the MDR1

gene was placed upstream from an IRES that allowed coexpres-

sion of a human glucocerebrosidase eDNA, efficient expression

of both genes was also seen (29).

The pHa-MDR-IRES-TK construct is able to confer multi-

drug resistance when it is transfected into drug-sensitive cells or

transduced as a retrovirus. However, in both transfection and trans-

duction experiments, numbers of drug-resistant colonies derived

from the pHa-MDR-IRES-TK were a few-fold lower than those

derived from pHaMDR. One reason for these lower frequencies

may be that pHa-MDR-IRES-TK has an insert 1.8 kilobase longer

than that of pHaMDR. Another reason may be that the downstream

sequences (IRES-TK) affect the transcription of the bicistronic

mRNA and/or the translation of the P-glycoprotein.

We examined the expression level of human P-glycoprotein in

pHa-MDR-IRES-TK-transduced, nonselected cells (Fig. 5). As

Research. on March 23, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

A B

C D

1(

Fluorescence Intensity

Clinical Cancer Research 455

a)

Eza)

0

a)-�

Eza)

0

Fig. 5 FACS analysis of nonseleeted cells transduced with pHa-MDR-IRES-TK or pHaMDR. The supernatant (2.5 ml) of retrovirus-producing cellswas used for transduction. Cells (106) were harvested after trypsinization, stained with the anti-P-glyeoprotein mAb MRK16, washed, and stained with

fluorescein-conjugated antimouse IgG. The fluorescence was analyzed by FACS. In each fluorogram, the single fluorescence peak of the parental cells

(left) was superimposed. MRK16-positive cells are shown as a small fluorescence-positive peak (right). A, Ltk cells transduced with pHa-MDR-IRES-TK (supernatant 1); B, Ltk cells transduced with pHaMDR; C, NIH3T3 cells transduced with pHa-MDR-IRES-TK (supernatant 1); D, NIH3T3

cells transduced with pHaMDR.

Table 6 Correlation of P-glyeoprotein-positive cells and HAT-

resistant cells in pHa-MDR-IRES-TK-transduced Ltk populations

P-glycoprotein- HAT-

Retrovirus positive cells” resistant cells”

supernatant (%) (%)

pHa-MDR-IRES-TK (ml)Supernatant 1 1 7.! 6.1

2.5 14.6 12.6

Supernatant 2 1 4.5 4.8

2.5 7.5 11.4

pHaMDR (ml) 1 8.4 0

2.5 18.2 0

a Percentage of MRK16-positive cells determined by FACS in

MDRI retrovirus-transdueed, nonseleeted population.

b Percentage of HAT-resistant cells in MDR1 retrovirus-transduced

population.

shown in Fig. 5, the mean fluorescence intensities of the pHa-

MDR-IRES-TK-transduced cells was almost the same as those of

the pHaMDR transfectants. Because these cells have not been

treated with any drugs, the expression level of P-glycoprotein

Table 7 Transduction efficiency of pHa-MDR-IRES-TK and

pHaMDR in NIH3T3 and Ltk cells

Retrovirus

supernatant

P-glyeoprotein- positive cells” (%)

Ltk NIH3T3

pHa-MDR-IRES-TK (ml)Supernatant 1 1 7.1 8.9

2.5 14.6 18.7Supernatant 2 1 4.5 6.0

2.5 7.5 13.2

pHaMDR (ml) 1 8.4 12.82.5 18.2 24.9

a Percentage of MRK16-positive cells determined by FACS.

directly reflects the ability of the retroviral constructs to drive the

expression. Therefore, we can conclude from this result that pHa-

MDR-IRES-TK confers similar levels of multidrug resistance as

pHaMDR when the vectors are introduced into drug-sensitive cells.

Research. on March 23, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

456 Retroviral Coexpression of MDR1 and HSV-TK

This result along with the observation that the pHa-MDR-IRES-TK

construct produces somewhat fewer vincristine-resistant colonies

than pHaMDR indicate that the true difference between the vectors

is relatively small.

The HAT-selected pHa-MDR-IRES-TK transfectants

showed significant levels of resistance to vincristine although

the cells had not been treated with vincristine, and all of the

transfectants expressed human P-glycoprotein by FACS analy-

sis using MRK16. When we establish MDR1-transfected cells,

we usually use antitumor agents such as vincristine or Adria-

mycin to select the MDR1-expressing cells. It is possible that

certain mechanisms of drug resistance could be activated or

selected for in recipient cells during drug treatment. In this

IRES-TK system, we can establish MDR1-expressing cells with

high efficiency without treating cells with multidrug resistance-

related drugs. This strategy is useful to express MDR1 mutants

whose activity to confer drug resistance is unknown or to

analyze the cross-resistance patterns of MDR1 mutants. Such

experiments are now ongoing in our laboratory. Apart from

studies on P-glycoprotein itself, the pHa-MCS-IRES-TK con-

struct should prove useful for the expression of many different

genes in TK-deficient cells to analyze the function of these

genes without needing to select directly for them.

When we compared drug resistance among the HAT-se-

lected clones and vincristine-selected clones (Tables 2 and 3),

the vincristine-resistant clones seemed to have a higher resis-

tance to vincristine than the HAT-selected clones, when 35

ng/ml vincristine was used for the selection. This stringent

selection condition could inhibit the growth of some transfec-

tants (Table 3). Low level vincristine-resistant clones, such as

might be selected in HAT, would not grow in this concentration

of vincristine. Therefore, the transfectant clones obtained after

vincristine selection might have higher vineristine resistance on

the average than HAT-selected cells. On the other hand, there

was no difference in HAT resistance between the HAT-selected

clones and vincristine-selected clones. HAT medium has no

effect on the growth of most pHa-MDR-IRES-TK transfectant

clones (Table 3). This result suggests that the selection is less

stringent for HAT resistance than for vincristine resistance, i.e.,

that some of the cells express relatively low amounts of the

bicistronic message and P-glycoprotein, but express enough

HSV-TK to allow growth in HAT medium.

One of the main goals of this study was to kill the MDR1-

transduced cells selectively using gancicbovir. For this purpose,

it is desirable that all of the MDR1-transduced cells express the

HSV-TK gene. To clarify this point, we examined the proba-

bility that expression of P-glycoprotein would not be accompa-

nied by expression of HSV-TK. When we analyzed the growth

of 20 randomly isolated vincristine-resistant clones and 20

HAT-resistant clones from pHa-MDR-IRES-TK-transfected

Ltk cells, all of the HAT-resistant clones (20/20) showed more

resistance to vincristine than the parental Ltk and 19 vincris-

tine-resistant clones of 20 (19/20) showed HAT resistance. Only

one clone (termed L39) showed HAT sensitivity (Table 2). The

overall frequencies of gancicbovir-resistant cells in a larger

vincristine-resistant, pHa-MDR-IRES-TK-transfected cell pop-

ulation were 2.2-3.1% for ‘I’-cre cells and 1.4-1.7% for Ltk

cells (Table 4). Not surprisingly, the gancicbovir-resistant cells

were shown to express human P-glycoprotein (Fig. 4). These

results suggest that 1-5% of the pHa-MDR-IRES-TK-trans-

fected cells may express the MDR1 gene alone and be resistant

to gancicbovir. Therefore, it is possible that there will remain a

small but potentially troublesome population of multidrug-re-

sistant tumor cells that will not be killed by this strategy. In a

previous report, treatment with ganciclovir in vivo against s.c.

injected tumor in mice produced complete tumor regression

when 50% of the tumor cells were transduced with the HSV-TK

retrovirus (1 1). The mechanism mediating the killing of by-

stander untransduced tumor cells is not fully understood; how-

ever, this result suggests that the minor population of the pHa-

MDR-IRES-TK-transfected cells expressing the MDR1 gene

alone can also be killed by ganciclovir treatment in vivo. Animal

studies would be needed to test this hypothesis.

Another possible problem with this strategy is that ganci-

cbovir will kill normal bone marrow cells intentionally trans-

duced with pHa-MDR-IRES-TK as well. In clinical situations,

patients will be treated with ganciclovir only when MDR-

transduced tumors present as disease. Since this may occur

several months to years after the gene therapy and subsequent

cancer chemotherapy, normal bone marrow cells that were not

transduced with the retrovirus may occupy the bone marrow of

the patients. Clinical studies examining the expression of the

MDR1 gene as well as the HSV-TK gene in bone marrow cells

for long periods of time would resolve this issue.

These preliminary studies suggest that the pHa-MDR-

IRES-TK safety-modified vector may be useful for gene therapy

of human cancer to protect bone marrow cells during intensive

chemotherapy. Further preclinical studies in various animal

model systems are needed to prove this hypothesis.

ACKNOWLEDGMENTS

We thank Drs. T. Tsuruo and K. Ono for kindly providing the

MRK16 antibody, Dr. R. Mulligan for ‘I’-ere and ‘1’-crip cells, and

Dr. A. Johnson for help with the Northern blot analysis.

REFERENCES

1. Chen, C. J., Chin, J. E., Ueda, K., Clark, D. P., Pastan, I., Gottesman,

M. M., and Roninson, I. B. Internal duplication and homology withbacterial transport proteins in the mdrl (P-glyeoprotein) gene from

multidrug-resistant human cells. Cell, 47: 381-389, 1986.

2. Ueda, K., Clark, D. P., Chen, C. J., Roninson, I. B., Gottesman,

M. M., and Pastan, I. The human multidrug-resistanee (eu/ri) gene:eDNA cloning and transcription initiation. J. Biol. Chem., 262: 505-

508, 1987.

3. Gottesman, M. M., and Pastan, I. Biochemistry of multidrug resis-

tanee mediated by the multidrug transporter. Annu. Rev. Biochem., 62:

385-427, 1993.

4. Ueda, K., Cardarelli, C., Gottesman, M. M., and Pastan, I. Expres-

sion of a full-length eDNA for the human ‘ ‘MDR1 “ gene confers

resistance to colehicine, doxorubicin, and vinblastine. Proc. Natl. Acad.

Sci. USA, 84: 3004-3008, 1987.

5. Pastan, I., Gottesman, M. M., Ueda, K., Lovelace, E., Rutherford, A.,

and Willingham, M. C. A retrovirus carrying a MDR1 eDNA confers

multidrug resistance and polarized expression of P-glyeoprotein in

MDCK cells. Proc. Natl. Acad. Sci. USA, 85: 4486-4490, 1988.

6. Chaudhary, P. M., and Roninson, I. B. Expression and activity of

P-glyeoprotein, a multidrug efflux pump, in human hematopoietie cells.

Cell, 66: 85-94, 1991.

7. Sorrentino, B. P., Brandt, S. J., Bodine, D., Gottesman, M. M.,

Pastan, I., Cline, A., and Nienhuis, A. W. Retroviral transfer of the

Research. on March 23, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

Clinical Cancer Research 457

human MDR1 gene permits selection of drug resistant bone marrow

cells in vivo. Science (Washington DC), 257: 99-103, 1992.

8. Podda, S., Ward, M., Himelstein, A., Richardson, C., Fbor-Weiss, E.,Smith, L., Gottesman, M. M., Pastan, I., and Bank, A. Transfer and

expression of the human multiple drug resistance gene into live mice.

Proc. Natl. Acad. Sci. USA, 89: 9676-9680, 1992.

9. Galski, H., Sullivan, M., Willingham, M. C., Chin, K. V., Gottesman,

M. M., Pastan, I., and Merlino, G. T. Expression of a human multidrugresistance eDNA (MDR1) in the bone marrow of transgenic mice:

resistance to daunomycin-induced leukopenia. Mob. Cell. Biol., 9:

4357-4363, 1989.

10. Mickiseh, G., Licht, T., Merlino, G. T., Gottesman, M. M., andPastan, I. Chemotherapy and chemosensitization of transgenie micewhich express the human multidrug resistance gene in bone marrow:

efficacy, potency, and toxicity. Cancer Res., 51: 5417-5424, 1991.

1 1. Culver, K. W., Ram, Z., Wallbridge, S., Ishii, H., Oldfield, E. H.,

and Blaese, R. M. In vivo gene transfer with retroviral vector producer

cells for treatment of experimental brain tumors. Science (Washington

DC), 256: 1550-1552, 1992.

12. Ram, Z., Culver, K. W., Walbridge, S., Blaese, R. M., and Oldfield,

E. H. In situ retroviral-mediated gene transfer for the treatment of brain

tumors in rats. Cancer Res., 53: 83-88, 1993.

13. Emerman, M., and Temin, H. H. Genes with promoters in retrovirusvectors can be independently suppressed by an epigenetic mechanism.Cell, 39: 459-467, 1984.

14. Germann, U. A., Gottesman, M. M., and Pastan, I. Expression of amultidrug resistance-adenosine deaminase fusion gene. J. Biol. Chem.,

264: 7418-7424, 1989.

15. Germann, U. A., Chin, K. V., Pastan, I., and Gottesman, M. M.Retroviral transfer of a chimeric multidrug resistance-adenosine deami-

nase gene. FASEB J., 4: 1501-1507, 1990.

16. Kaufman, R. J., Davies, M. V., Wasley, L. C., and Michnick, D.Improved vectors for stable expression of foreign genes in mammaliancells by use of the untranslated leader sequence from EMC virus.

Nucleic Acids Res., 19: 4485-4490, 1991.

17. Morgan, R. A., Couture, L., Elroy-Stein, 0., Ragheb, J., Moss, B.,and Anderson, W. F. Retroviral vectors containing putative internalribosome entry sites: development of a polycistronic gene transfer

system and applications to human gene therapy. Nucleic Acids Res., 20:

1293-1299, 1992.

18. Dirks, W., Wirth, M., and Hauser, H. Dicistronic transcription units

for gene expression in mammalian cells. Gene, 129: 247-249, 1993.

19. Sugimoto, Y., Aksentijevich, I., Gottesman, M. M., and Pastan, I.Efficient expression of drug-selectable genes in retroviral vectors undercontrol of an internal ribosome entry site. Biotechnology, 12: 694-698,1994.

20. Danos, 0., and Mulligan, R. C. Safe and efficient generation ofrecombinant retrovirus with amphotropic and ecotropic host ranges.

Proc. Natl. Acad. Sci. USA, 85: 6460-6464, 1988.

21. Miller, A. D., and Buttimore, C. Redesign of retrovirus packaging

cell line to avoid recombination leading to helper virus production. Mob.

Cell. Biol., 6: 2895-2902, 1986.

22. Sugimoto, Y., and Tsuruo, T. DNA-mediated transfer and cloning

of human multidrug-resistant gene of Adriamycin-resistant myeloge-

nous leukemia K562. Cancer Res., 47: 2620-2625, 1987.

23. Kioka, N., Tsubota, J., Kakehi, Y., Komano, T., Gottesman, M. M.,

Pastan, I., and Ueda, K. P-glycoprotein gene (MDR1) eDNA from

human adrenal: normal P-gbycoprotein carries Gly’85 with an altered

pattern of multidrug resistance. Biochem. Biophys. Res. Commun., 162:

224-231, 1989.

24. Chen, C., and Okayama, H. High efficiency transformation ofmammalian cells by plasmid DNA. Mob. Cell. Biol., 7: 2745-2752,

1987.

25. Hamada, H., and Tsuruo, T. Functional role for the 170- to 180-kDa glycoprotein specific to drug-resistant tumor cells as revealed by

monoclonal antibodies. Proc. NatI. Acad. Sci. USA, 83: 7785-7789,

1986.

26. Ishida, Y., Ohtsu, T., Hamada, H., Sugimoto, Y., Tobinai, K.,

Minato, K., Tsuruo, T., and Shimoyama, M. Multidrug resistance in

cultured human leukemia and lymphoma cell lines detected by a

monocbonal antibody, MRK16. Jpn. J. Cancer Res., 80: 1006-1013,

1989.

27. Chomczynski, P., and Sacchi, N. Single-step method of RNA iso-lation by acid guanidinium thiocyanate-phenol-chloroform extraction.

Anal. Biochem., 162: 156-159, 1987.

28. Thomas, P. 5. Hybridization of denatured RNA and small DNAfragments transferred to nitrocellulose. Proc. Natl. Acad. Sci. USA, 77:

5201-5205, 1980.

29. Aran, J. M., Gottesman, M. M., and Pastan, I. Drug-selected

coexpression of human glucocerebrosidase and P-glycoprotein using

a bicistronic vector. Proc. Natl. Acad. Sci. USA, 91: 3176-3180,

1994.

30. Johnson, A. C., Garfield, S., Merlino, G. T., and Pastan, I. Expres-sion of epidermal growth factor receptor proto-oncogene mRNA in

regenerating rat liver. Biochem. Biophys. Res. Commun., 150: 412-

418, 1988.

31. McLachlin, J. R., Eglitis, M. A., Ueda, K., Kantoff, P. W., Pastan,

I. H., Anderson, W. F., and Gottesman, M. M. Expression of a humancomplementary DNA for the multidrug resistance gene in murine he-

matopoietic precursor cells with the use of retroviral gene transfer. J.

Natl. Cancer Inst., 82: 1260-1263, 1990.

32. Shen, D-w., Cardarelli, C., Hwang, J., Cornwell, M., Riehert, N.,

Ishii, S., Pastan, I., and Gottesman, M. M. Multiple drug resistant human

KB carcinoma cells independently selected for high-level resistance to

cobehicine, adriamycin or vinblastine show changes in expression ofspecific proteins. J. Biol. Chem., 261: 7762-7770, 1986.

Research. on March 23, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

Research. on March 23, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

1995;1:447-457. Clin Cancer Res   Y Sugimoto, CA Hrycyna, I Aksentijevich I, et al.   selective killing of MDR1-transduced cellsbicistronic messenger RNA in a retrovirus vector allowsherpes simplex virus thymidine kinase gene as part of a Coexpression of a multidrug-resistance gene (MDR1) and

  Updated version

  http://clincancerres.aacrjournals.org/content/1/4/447

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/1/4/447To request permission to re-use all or part of this article, use this link

Research. on March 23, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from